Unassigned shards in elasticsearch

Question

We have a 0.90 cluster with two elasticsearch nodes. For a while now their status was going yellow often then coming back to green, but now it's red and seems to stay that way. The health is this :

{
  "cluster_name" : "name",
  "status" : "red",
  "timed_out" : false,
  "number_of_nodes" : 4,
  "number_of_data_nodes" : 2,
  "active_primary_shards" : 724,
  "active_shards" : 1448,
  "relocating_shards" : 0,
  "initializing_shards" : 0,
  "unassigned_shards" : 544
}

I googled a bit and found that I should list the unallocated shards, but _cat doesn't seem to exist in 0.90. Is there a way to get back to green ? Both nodes can ping each other of course, so I don't see a reason why it wouldn't work.

score 0 · Accepted Answer · answered Jun 14 '16 at 13:42

Found an answer on another post, here it is :

curl -XPOST 'localhost:9200/_cluster/reroute' -d '{ "commands": [{ "allocate": { "index": "my-index", "shard": 0, "node": "node_name", "allow_primary": 1 } }] }'

I had to do that for each shards of all the indexes needing re-allocating, there were 500 of them. That was just great. Anyway, it's solved.

Unassigned shards in elasticsearch

1 Answers1